Bezos’s law signals it’s time to ditch the data center

As more and more operational (e.g. first responders, security professionals, and emergency managers) consider the cloud, this article should be used as a reference


Wherever you stand on the debate over which cloud giant will reign supreme, it’s clear that the economic forces shaping the market are evolving quickly. After nearly three decades helping companies move their enterprise applications into the modern era, whether to new servers, operating systems or clouds, I’ve seen the cycle before: innovation leads to rapid expansion, which leads to consolidation, shake-out and more innovation. We’ve been down this road before.

Now comes new cloud computing data based on Total Cost of Infrastructure (TCOI), proving cloud providers are innovating and reducing costs in areas beyond hardware. The result is a more compelling case for cloud as a far cheaper platform than a build-your-own data center. Further, the economic gap that favors the cloud provider platform will only widen over time.

In many ways, cloud computing is bringing to the enterprise world what Henry Ford brought for cars. Ford developed and…

View original post 684 more words

General Thoughts and Rambles, GIS and Mapping, Situational Awareness

How Common is Our Common Operating Picture

I’ve been involved in establishing Common Operating Pictures (COPs) for about 15 years or more. The only definitive thing that I can tell you is that they don’t exist once you involve multiple parties (services, branches, agencies, etc.). Many attempts at creating a COP have been made but the reality is that each organization or user needs their own context to be applied so a picture simply can’t do it. Once the contexts change there is little common to the picture. Over time the term Common Operating Picture has left my vocabulary and I tend to get uptight when it is used (abused).

Some military organizations have used the phrase “Common Operating Environment” (COE) in recognition of the fact that your context determines what information is useful and the environment (systems) should adapt to your needs. A COE is nebulous though – there is no clear definition of what it is and how well it works. Other military groups came up with “Common Relevant Operating Picture” (CROP) but these both lose the idea. Imagine a military command centre with just a few basic positions:

  • INT (the 2)
  • Ops (the 3)
  • Logistics (the 4)

Can you seriously imagine that these three players need exactly the same picture? Doubtful. However, they do need to share relevant information. The Ops officer needs the INT and Logistics side of the house to understand the current situation well enough so they can bring their own information and expertise to bear on the situation. Once we add civilians to the mix the terminilogy and needs of the user’s picture really starts to morph.

However, there is an answer to this holy grail search for a COP – it’s in the data.

The data should be common – but unless you and I are trying to do exactly the same thing the picture likely isn’t common at all.

General Thoughts and Rambles, Information Exchange, Situational Awareness

Information Sharing – A New Liability?

I keep hearing discussions about the potential liability of sharing information. There is a concern that sharing something with someone else can expose a liability. In the public safety and Emergency management space this perceived liability comes up a lot.

This concern gets mentioned with respect to social media in particular but it applies to private sharing too.

I agree that there are definitely some issues with sharing information. When you share information, you need to make sure, to the best of your abilities under the situation that you are operating in, that what you share is accurate, timely, and relevant for your recipients. That applies whether you are sharing privately or publicly. So, if you have handled these criteria, and you have authority to release information why wouldn’t you?

Well, you’re likely paranoid about the aforementioned liability. But let me throw this prediction out:

At some point liability will lie upon a group that had information, could have shared it (it was accurate, timely, relevant, within their mandate, etc.) – but DID NOT. I really think that liability due to failure to share will be a tipping point for many organizations. Once someone is taken down for failing to share the reticence will drop and we’ll all be looking for the guidance on how we can share better – not whether we should share.

I’ll reiterate my blog disclaimer that my opinions are mine – not my customer’s. That being said I’m working with leaders across Canada and around the world to make sure that when everyone is ready to share we will have some structured guidance on how it should be done…

General Thoughts and Rambles, GIS and Mapping, MASAS, Situational Awareness

Keep the Machine Simple …

I was watching a very interesting TED video (Amos Winter: The cheap all-terrain wheelchair) about Amos Winter’s efforts to create a wheelchair for the developing world. His presentation and the apparent success of the program is very impressive. The engineering approach he uses applies to so much of the world and it started me thinking  about some of the efforts that we’re working on for MASAS.

Continue reading

General Thoughts and Rambles, MASAS, Uncategorized

PlayBook – First Impressions

I’m a big fan of unpacking new gadgets. I enjoy the first impressions that I get usually. I’ll admit that I find the unboxing experience of Apple products to be pretty amazing and I use them as the bar, knowing that is a high standard.

So, I started unboxing a new PlayBook (64GB) that I picked up for a project I am working on. The packaging is fairly good. A lot better than the weak efforts that Dell puts forth for example. Here are some early thoughts that come from setting it up (twice) and using it for a short period of time:

  • Account – I created my Blackberry ID and managed to munge my email address (@Xontinuumloop…) and that isn’t a good thing at all. To recover you have to finish the setup then do a full data wipe, which restarts everything, including re-doing the PlayBook OS update (500+MB).
  • Tutorials – These MANDATORY tutorials teach you about the Home Screen (swipe bottom edge up) and the Applications Menus (swipe top edget down). Given that I had to do a wipe I had the pleasure of having to do this twice – not fun.
  • Form Factor – The device is a good size though. I find my iPad (gen 3) is too big to be grabbing all the time. When I read at night I reach for my iPhone to check emails and now that I have this PlayBook, I may reach for it.
  • Apps – Apps are a huge weakness. The RIM folks have never understood apps like Apple and Microsoft do. They’ve always been very rigid in what they allow and unfortunately that is reflected by a dearth of apps. Even the baseline apps that come pre-installed aren’t very popular. Look at Zinio, which is installed. Go to Zinio’s site and you can’t even find mention that they have something other than iOS (iPad and iPhone/iPod touch variants), Android, and even WebOS are on the products page. No mention of the PlayBook app.

There are some really good things coming for BB10 and I will say that I am impressed with RIM opening up on the development front. However, with last weeks quarterly results, yet another delay on BB10, and the general malaise in the RIM market I’m worried. RIM used to be pretty damned hot and the problem I think is that they knew it.

I’m looking forward to doing some MASAS work on the PlayBook. The HTML JavaScript extensions that RIM added for capturing Audio, Pictures, and the high level of HTML5 support make it look pretty powerful. The concepts will be portable though. I just hope that RIM is still around long enough to get out of this funk they created.


Design – Critical everywhere

Here’s a great video if you’re involved in the rollout of a product – it covers design largely, but delves into the team dynamics as well

Some key takeaways.

  • Business often cuts design out, it’s easy to put on the “chopping block”
  • People don’t want to pay for design
  • Design is never an accident
  • Standards are a great starting point – think about them as rules …
  • “The Rules” : learn them, follow them, study them, then break them.
  • “Ideas are cheap – execution is hard”
General Thoughts and Rambles, Standards

Old Rigid Standard – Almost Ate My First Company

Back in … ahem … the late 90s, my partner and I made our first international sale. We were juiced – all we had to do was hook up the search and rescue software that we had (SARMaster) to the US Mission Control System – the system that manages the US segment of COSPAS-SARSAT. Simple – right? We’d already done it in Canada so how different could it be?

We price our bid low – I recall it was well under $250K (it may have been <$100K) for the full site. When we got there, ready to write our final bits of software in the hotel (love Courtyard Marriott!) we were in for one heck of a surprise when we tried to hook up to the AFTN. AFTN (Aeronautical Fixed Telecommunications Network) was used to do air-traffic control so it had to be rigid and controlled. It wasn’t at all ready for what a couple of new university grads were trying to do with it. It has been upgraded since then, but it is still a pretty strange and archaic beast from what I can tell. It’s a standard – but it is an old, rigid, and effectively closed (you pay for entry) network – no newbies allowed.

We reached out and found out that the development kit and testing kit for hooking up to the AFTN was more money than we were charging for the full system. I can’t remember if the dev kit was $250K or not, but I remember it was insanely expensive and would have bankrupted us out of the gate.

So – what’s a 3-man company (may have been 4) to do? Well, buying the kit was out of the question so we started Macgyver-ing. The AFTN terminal at AFRCC terminated at a serial printer.

So – how do you teach a computer to pretend it is a serial printer – well, first off you get a serial guy (we had one – he used to do Cirque de Soleil stage controls), then you teach the computer to play stupid at its end (printers are quite stupid in RS-232 speak). To do this you basically ignore all of these computer-centric RS-232 guidance and start ignoring some of the signals – who needs an ACK?

Once we figured out that we could emulate a printer, we were off to the races – I’ll save some stories for later, but the late nights were insane.

The staff at AFRCC were incredible – silly buggers thought we were sleeping at times – we were still there when they returned for their shifts. After a two-week crush we had the system working.

The AFRCC controllers had one complaint – the new system didn’t make any noise.

So, we added sound – they picked it …

… it sounded like a dot-matrix serial printer.

General Thoughts and Rambles, Information Exchange, Standards

Jeep – Proof of Standards

I’m big into the use of structured information, particularly for information exchange and in getting multiple groups to share. There are loads of excellent reasons for the use of structured information but quite often new formats and schemas are created on the fly, resulting in a massive amount of work for everyone. The successful systems I have been involved with or observed have used standards to ensure that everyone uses the same formats. I’ll ramble on about that in the future, as I’m passionate about it. For now though I want to throw out an analogy that I kind of stumbled upon.

A few months back I started looking for a new vehicle (the current one was getting older and would do weird things like spontaneously open its sunroof). I looked around at various vehicles and stumbled upon the Jeep Wrangler while looking at the Grand Cherokee. I had no plans to get one.

Then I went home and did some research.

I joined multiple forums to see what people thought of the various Jeep models. One forum (www.jeepforum.com) is broadly Jeep and has 324,000+ members. The Wrangler forum (www.wranglerforum.com) has almost 90,000 members. I thought that the numbers were artificially high – then I noted that between those two forums there are over 14,000,000 posts in over 1.1 million threads – that’s a very, very active community.

I did some more reading and realized that the Wrangler folks on both forums were the most rabid and verbose. Typically the discussion was about modifications (mods) that have been done, are being done, or will be done. That seemed pretty heavy to me: why is everyone modding their vehicle?

From the number of modifications being done (literally hundreds per day often) I realized that something was different about the Wrangler platform. I’ve used platform here in the form that Detroit uses it, but that will change before the end of this article.

Let’s take two areas where many different modifications occur:

  • Bumpers – there are kits ranging from the very basic (plastic cover that are put on after lopping of the ends of the front bumper) to the very heavy and complex (full integration of lights, winch, recovery points, bush/bull/spider bars, etc.)
  • Lift Kits – there are budget boosts (BB), body lift kits, suspension lift kits, and more – all ranging in the lift height (0.5″, 1.0″, 1.5″, 2.0″, 2.5″, – you get the picture)

Under both categories I can find at least 20 large vendors that provide solid kit over a vast range of prices. When you factor in the small fabrication shops that are very local the bumper market alone must have hundreds of vendors in North America alone.

I started wondering what could possibly warrant such a large ecosystem of manufacturers. What was it that made so many companies jump in? They’ve been around, in many cases, for decades – so this isn’t a fad.

Then it donned on me – the Wrangler platform is a STANDARD. My model is a JKU. The major standards/platforms of the past few decades have been the JK/JKU, TJ, YJ, and CJ (newest to oldest).

What does a platform mean it is when considered it a standard? It means that a manufacturer can build products based on it – there won’t be massive changes made for some period time. The Jeep standard only goes through major changes (new platform) about every 10 years or so. This means a manufacturer can plan a product line, based on a known standard, that has life to it.

Sure, there are slight changes through the standard lifecycle, much like in the information standards I work with. Let’s take the current platform (JKU) bumpers and compare this to the OASIS Common Alerting Protocol which I work with.

CAP started out as version 1.0 (2004) and has evolved slowly (1.1 in 2006, 1.2 in 2010). There haven’t been major changes yet but adjustments have been required by the open-source and vendor community that supports CAP.

The JK/JKU platform hasn’t changed wildly since it’s release in 2007, but there have been changes. Lately for example, many bumper manufacturers have had to make recommendations to their customers about moving a vacuum pump that showed up in a new location when Jeep changed the engine. Think about that – the full engine, and all of its supporting parts and functions, was changed – and the net result is that people had to learn where to move the vacuum pump to accomodate the standards-based bumper. In essence a mini-standard (a standards “Profile” is what we would call this in OASIS) was created, on-the-fly, by the industry – and multiple manufacturers are recommending the same approach.

That’s just one area – bumpers. The lift kits, the roof covers, the seat covers, and many more sub-industries have grown out of the Jeep Wrangler standard. The known rate of change and the known specifications of the Wrangler standard has the same industry effect as the information exchange standards that I see all the time. Vendors and open-source applications don’t have to go and reinvent things.

I’ll leave a future topic for the minor adjustments that standards require to “fit” properly, and I may use a Jeep analogy there too (some bumper installs require a bit of grinding for optimal fit apparently).

For now, I’ve rambled on too long – I’ll likely revisit many of these topics over time and dig a little deeper.

Oh – Wrangler owners have another standard – they wave at each other on the road (the Jeep Wave). It’s not a hard and rigid standard (some don’t even do it) and it has various permutations on technique (two finger lift from wheel, full hand wave, wave out window, and the best – wave out the roof). It is definitely widespread though. There’s an addendum to the Jeep Wave standard – you have to smile after you do it…

Situational Awareness

We deal on a national level at the moment – sharing information from agency to agency. The MASAS project is making great progress and generating a lot of buzz. Some would say, and they have, that we have “nailed it”.

To me, “nailed it” means that a job is done. In the case of MASAS and the broader MASAS ecosystem, it means that the job is just getting started.

MASAS is focused right now on inter-agency information sharing. A lot of that activity is based on pushing key reports (SITREPs, notifications, requests). Though we have interest from tri-services in many places they deal with a level of tactical SA that is truly amazing.

Take a rip through this Tracking of Personnel article to get an idea of how tough maintaining tactical SA can get.

Rich Casaway (bio and blog here) is hard core first responder. He’s a leader in the first responder community – over 30 years of first responder experience and a PhD focused on SA. His blog provides a very impressive and passionate guide for maintaining high levels of situational awareness.

I’m particularly amazed at the depth of thought that has gone into the comms guidance – in 13 words a radio briefing can provide perfect SA on the activities of a small crew:

RADIO: Captain 1, with 3, on Attack in the basement. Conditions are not improving.

  • Providing the Who (captain plus crew of 3), Where (basement), Why (on Attack), and What (update on conditions). As Rich calls it: Personal identifier + crew size + location + update report

Impressive …

Tactical SA – Radio Comms – Tracking Personnel

General Thoughts and Rambles, Situational Awareness

Collaborate or Perish

I was reading through Eric Holdeman’s commentary about a new book (Collaborate or Perish – disclosure: Amazon.ca affiliate link) – though he, like me hasn’t read the full book – there is a Q&A with one of the authors and he knows the players.

Eric disagrees with many of the broad-brush statements raised in the Q&A (follow Eric’s blog to the Q&A). He then brings up a good approach that collaboration can only occur after trust has been established – at the individual and organization level. He is correct that establishing this trust takes time – a lot more time than one would think.

Eric’s point about old baggage isn’t lost on the work that I’m lucky enough to be doing with the MASAS team.  We’ve seen barriers fall down in front of us – between groups that we had assumed were working together. Groups that may even have been hostile in the past (baggage) are working together and establishing those critical individual and organization-level relationships. It took a while for us to realize why this is – but Eric’s point about a possible solution, is one that I would say is a mandatory – a “third party neutral party” must be involved.

To me the third party must be there – any system of collaboration that is hosted by a single entity, as good/great as that entity is – will be seen as “their system”. Some groups will join, but many will stand back and stay out for various reasons (past relationships, information ownership, and many others). Others will actively campaign against it either to put forward their own solution or to simply fight the idea.

What we have seen with MASAS is that information exchange will only begin after creating that neutral third-party that isn’t run by any one particular interest – this may be influenced and funded by groups, but it is its own entity with a mandate of neutrality. Further, it needs transparency in its operations and governance. Trust is a hard thing – it takes time to build and it builds slowly. So, to make things work you can’t rush it and you can’t skip steps.

Establishing situational awareness is tough. It gets tougher when multiple agencies are involved. The dynamic of what information is needed changes as each agency has its own focus and context. Adding a trust barrier makes it only harder. Working collaboratively requires that trust but being shown a “trust me – I’m big” type of system isn’t going to get us through that trust barrier.